The best books on scientific method: how science works, fails to work and pretends to work

Richard Feynman was unique. A brilliant theoretical physicist, humorous, eccentric, and independent.

Feynman’s genius gave him a certain freedom, which he exploited to the full. The book is autobiographical and shows his often irresponsible behavior but also a relentless curiosity, and willingness to try anything, the essence of a successful scientist.

One cannot hope to imitate Feynman (and perhaps we should not: he was often mischievous, even mildly malicious); but any scientist should envy the way he approached problems in engineering as well as science—and the book is fun!

James Watson was a clever, pushy, and critical young American molecular biologist exposed to the scientific culture of Britain in the early 1950s.

The book is full of acerbic comments about “stuffy” Cambridge dons and the rules of etiquette that young Jim struggled with, all the while scheming to maintain the various fellowships that allowed him to remain in the UK and pursue his ambition: to understand the chemical nature of the genetic material, DNA.

The book provides a lively account of his collaboration with an older Brit, the brilliant Francis Crick, who was also trying to unscramble DNA. Much of the technical stuff will be incomprehensible to most, but the method the two followed is clear. The partnership was hugely fruitful and the book is a lively account of how science actually works.

Watson and Crick tried everything while coping with competitors and their criticisms as well as their findings, which as often contradicted as supported their speculations.

In the end, of course, they won.

Stuart Ritchie is a cognitive psychologist, a Lecturer at the University of London’s King’s College. A few years ago, he had an experience that seems to have been the impetus for this lively and important book.

A famous social psychologist published a paper in a respected scientific journal purporting to demonstrate extra-sensory perception. The method was conventional—he tested a group of subjects and averaged the results, even though the claimed ability should have been demonstrable by individuals. Averaging across a group made no sense.

The results turned out to be unreplicable (no surprise!), but the paper (by Ritchie and others) showing so was refused publication. Houston, we have a problem! And so does science as Ritchie shows in this thoughtful, well-researched, and surprisingly readable book on a difficult but hugely important topic.

Science, and the freedom of inquiry on which it depends, is at the heart of Western civilization. Science Fictions illustrates the dark side of modern science, in contrast to the open-minded brilliance of Feynman, Watson, and Crick and their ilk. Perhaps the contrast will make us all wiser. 

Experimental Medicine is a classic, a clear and simple account of physiological research, well-illustrated with comprehensible examples (no molecular biology needed).

It is well worth reading if only because so many of the principles that French physician Claude Bernard pioneered have been largely forgotten. Most important is his demand for certainty: “Science permits no exceptions,” an injunction now more often forgotten than obeyed. Bernard would have been appalled by, for example, a famous choice experiment in which 60% of a large group of subjects choose a less-profitable option.

The experimenters concluded that individual humans (not just groups) show risk aversion, ignoring the 40% who did not. In a similar situation, Bernard would always seek the conditions that yield a 100% result. He gives an example where he was able to resolve uncertainty concerning the sensitivity of spinal roots.

We’re still waiting for any serious effort to achieve a Bernardian certainty about human choice. But his book may encourage readers to make the effort.  

Any list of books about science must have something about Darwin. The book to read is The Origin. But that’s obvious. I don’t need to go into it. So here is a Darwin book that is less obvious.

Most scientists will never have heard of it; it’s literary history, not science. I learned of it by accident, in a talk given by an economic historian. This is a fascinating book by a literary historian from which I learned much about Darwin. Gillian Beer recognizes that Darwin was a great writer. She traces similarities between his rhetorical style and the strategies of some other iconic Victorian writers, such as George Eliot (Middlemarch) and Samuel Butler (The way of all flesh).

She also discusses scientific writers of the era such as George Lewis and Herbert Spencer. There are many revealing long quotations. I learned much from the book, and so will you.